Large-eddy simulation of enhanced mixing with buoyant plumes

Abstract

Bubble-induced turbulence is known to enhance mixing and diffusivity by increasing agitation in the continuous phase. This is commonly used in practical applications such as chemical reactors or aerators in wastewater treatment plants, yet there are few modelling tools capable of assisting the detailed design of these devices. In this work, large-eddy simulations of turbulent mixing produced by buoyant plumes and screens are analysed to assess the efficiency of bubble reactors. The solver uses an Eulerian–Lagrangian point-particle algorithm to provide two-way coupling of the liquid and gas phases. The transport of a passive tracer is simulated to quantify the homogeneity of the mixture and the speed of the process. The accuracy of the solver predicting mixing times is successfully validated versus experimental data in a bubble reactor with three different diffuser configurations and the sensitivity of the prediction to different bubble size distributions and mesh resolutions is analysed. The numerical results quantify the mixing induced by bubble screens and different arrangements of discrete plumes and explore the influence of the gas flow rate and the depth of the reactor. Our predictions show significant differences regarding mixing times and energetic efficiency for different aerator setups. For a constant flow rate, bubble screens provide a better performance than combinations of plumes. Our simulations also predict that arranging the gas diffusers as a screen is remarkably more efficient than increasing the gas flow rate.